System and method for sea cucumber peptide quality judgment and standardized application

Abstract

The invention relates to the field of sea cucumber peptide quality, and discloses a system and a method for sea cucumber peptide quality judgment and standardized application, which are used for performing multi-analysis window division on a sample area determined in a spectral absorption diagram and a microstructure image of a sea cucumber peptide sample; identifying an abnormal region in the spectrum curve based on a change gradient model, and removing an analysis window corresponding to the abnormal region in the sample region to obtain a candidate analysis region; judging whether the chemical component characteristics of the candidate analysis area are similar to the main component characteristics of the sea cucumber peptide sample or not through component deviation analysis; if yes, corresponding microstructure image data in the candidate analysis area are extracted, and particle morphological characteristics are calculated; and judging whether the morphological characteristics of the particles meet set evaluation requirements or not, if so, defining the candidate analysis area as an effective analysis area, and based on the spectral response data of the effective analysis area and the morphological characteristics of the particles. The method has the advantages of high throughput and automation.

Description

Technical Field

[0001] The present invention relates to the field of sea cucumber peptide quality, and in particular to an application system and method for sea cucumber peptide quality determination and standardization. Background Art

[0002] Sea cucumber peptides, marine functional peptides with high nutritional value and diverse biological activities, are currently widely used in health supplements, food additives, and functional medicines. With the rapid development of the sea cucumber peptide industry, the demand for product quality control and standardization is increasing. However, existing quality assessment methods are mostly based on low-throughput laboratory-based detection technologies, such as UV spectrophotometry, high-performance liquid chromatography, gel permeation chromatography, and amino acid composition analysis. These methods often have long detection cycles and complex procedures, requiring high operator expertise. Furthermore, experimental results are susceptible to human interference, making objective and stable assessment difficult. Especially in industrialized and large-scale production scenarios, existing testing processes cannot meet the demand for real-time, continuous, and rapid assessment of raw materials, intermediates, and final products. Furthermore, existing equipment has a low degree of automation, and the data collection and analysis process relies on manual intervention, resulting in low efficiency and accuracy. This lack of intelligent and standardized support severely hinders the development of the sea cucumber peptide industry towards high-quality, traceable quality. Therefore, the design of a high-throughput, automated system and method for sea cucumber peptide quality assessment and standardization is essential. Summary of the Invention

[0003] In response to the shortcomings of the existing technology, the present invention provides a system and method for determining and standardizing the quality of sea cucumber peptides, which has the advantages of improving detection efficiency and accuracy, meeting the needs of industrial quality control, and solving the problems in the above-mentioned background technology.

[0004] In order to achieve the above-mentioned purpose of improving detection efficiency and accuracy and meeting the needs of industrial quality control, the present invention provides the following technical solution: a method for determining and standardizing the quality of sea cucumber peptides, comprising the following steps:

[0005] The sample area determined in the spectral absorption map and microstructure image of the sea cucumber peptide sample is divided into multiple analysis windows, and the spectral response data of the corresponding position in the spectral absorption map of the analysis window is processed by derivative operation to construct a change gradient model;

[0006] Identify abnormal areas in the spectral curve based on the variation gradient model, and remove the analysis window corresponding to the abnormal area in the sample area to obtain the candidate analysis area;

[0007] Through component deviation analysis, determine whether the chemical composition characteristics of the candidate analysis area are similar to the main component characteristics of the sea cucumber peptide sample; if similar, extract the corresponding microstructure image data in the candidate analysis area and calculate the particle morphology characteristics;

[0008] Determine whether the particle morphological characteristics meet the set evaluation requirements. If so, define the candidate analysis area as the effective analysis area. Based on the spectral response data and particle morphological characteristics of the effective analysis area, construct a quality grading model, output the comprehensive quality score, and generate the quality standardization vector of the sea cucumber peptide sample.

[0009] Preferably, the process of performing derivative operation and constructing the change gradient model is as follows:

[0010] The sample area is divided into small windows of uniform size and numbered respectively. Each window has a unique spatial coordinate.

[0011] The corresponding spectral absorption curve is extracted for each window, and the first-order derivative processing is performed on each spectral curve using the Savitzky–Golay smoothing derivative algorithm;

[0012] The rate of change of each window is mapped back to the original image space coordinates to construct a two-dimensional change gradient model.

[0013] Preferably, the analysis window process corresponding to the abnormal area is removed from the sample area as follows:

[0014] The spectral curve comes from the collection of the spectral absorption diagram of the sea cucumber peptide sample, and the spectral absorption diagram is obtained by measuring the sea cucumber peptide sample;

[0015] Based on the constructed variation gradient model, a statistical judgment method is used to identify spectral abnormal areas and eliminate the corresponding analysis windows in the sample area;

[0016] The average change rate and standard deviation are calculated based on the spectral derivative change rate of all analysis windows, and the judgment threshold for spectral change anomaly is set. The judgment threshold is the weighted calculation value of the window change rate mean and standard deviation.

[0017] Preferably, the process of determining whether the chemical composition characteristics of the candidate analysis area are similar to the main component characteristics of the sea cucumber peptide sample is:

[0018] Based on the spectral response data of sea cucumber peptide samples, principal component analysis was used to extract the principal component eigenvectors, which were used as reference templates for chemical composition determination.

[0019] Aggregate and extract features from the spectral data of each analysis window in the candidate analysis area to form a feature vector of the candidate area;

[0020] The aggregated spectrum feature vector of the candidate analysis area and the principal component feature template vector of the sea cucumber peptide sample are input, and the two vectors are input into the similarity analysis model.

[0021] Preferably, the process of judging whether the particle morphology characteristics meet the set evaluation requirements is as follows:

[0022] Extracting raw image data from the corresponding microstructure image in the candidate analysis area, performing preprocessing operations, and obtaining particle contours;

[0023] Use image segmentation algorithm to extract particle area and number each particle;

[0024] Aggregate the morphological features of multiple particles in the region to form a morphological feature vector of the candidate analysis region;

[0025] Compare with the preset morphological evaluation requirements to determine whether all morphological indicators meet the set conditions.

[0026] Preferably, the process of generating the quality standardization vector of the sea cucumber peptide sample is:

[0027] For the regional data identified as the effective analysis area, the spectral response curve and particle morphological parameters are extracted respectively, and all features are normalized and spliced into a fusion feature vector;

[0028] Extract the fusion feature vector and the corresponding true quality score, build a training set, and use a supervised learning algorithm to model the sample data to obtain a scoring model;

[0029] The fusion feature vector extracted from the effective analysis area of the sea cucumber peptide sample to be tested is input into the model, and the comprehensive quality score is output;

[0030] A quality normalization vector is constructed based on the output comprehensive quality score and the corresponding weight contribution of each feature in the model.

[0031] A system for determining and standardizing the quality of sea cucumber peptides, comprising:

[0032] Multi-analysis window construction module: multi-window division is performed on the spectral absorption map of the sea cucumber peptide sample and the sample area selected in the microstructure image, and derivative operation is performed on the spectral response data;

[0033] Abnormal area identification module: identifies abnormal areas in the spectral curve based on the change gradient model, and eliminates the corresponding analysis window to obtain the candidate analysis area;

[0034] Morphological feature analysis module: determines whether the chemical composition characteristics of the candidate analysis area are similar to the main component, and extracts the particle morphological characteristics when the conditions are met;

[0035] Quality modeling module: Based on the spectral response data and particle morphological characteristics of the effective analysis area, a quality grading model is constructed, which outputs a comprehensive quality score and generates a quality standardization vector.

[0036] Compared with the prior art, the present invention provides a system and method for determining and standardizing the quality of sea cucumber peptides, which has the following beneficial effects:

[0037] 1. Divide the spectral absorption graph of the sea cucumber peptide sample into windows and extract the spectral response data of each window. Calculate the spectral change rate through derivative operation and construct a change gradient model that represents the intensity of the spectral response. Set an abnormal change rate threshold based on statistical characteristics, and determine the analysis window with a change rate exceeding the threshold as an abnormal area and eliminate it. Finally, retain the remaining windows to form a candidate analysis area with stable spectral response. This technical solution can effectively eliminate spectral abnormality areas in sea cucumber peptide samples caused by impurity interference, abnormal composition or uneven structure, significantly improve the composition consistency and data stability of subsequent regional analysis, and provide a high-quality, repeatable analysis basis for the construction of a sea cucumber peptide quality grading model, thereby improving the accuracy and standardization of quality evaluation.

[0038] 2. For the candidate analysis areas that have passed the composition deviation analysis, the image segmentation algorithm is used to extract the particle area, and each particle is numbered and the boundary is identified, and the morphological characteristic parameters of the particles are calculated; the various morphological parameters extracted are compared with the preset evaluation requirements one by one. If all morphological characteristic indicators meet the set qualification standards, the candidate analysis area is confirmed as a valid analysis area; based on the spectral response data of the valid analysis area and the corresponding particle morphological characteristic information, a sea cucumber peptide quality grading model is constructed and a comprehensive quality score is output to reflect the quality level of the sample, and a quality standardization vector with a universal expression is further generated for subsequent product comparison, classification and quality control management. The fusion evaluation of microstructure morphology and spectral response characteristics is effectively realized, so that the quality assessment of sea cucumber peptide samples no longer relies on a single dimensional indicator, thereby improving the scientificity and accuracy of quality grading; at the same time, through the output of the standardized vector, not only the reusability and traceability of the sample quality evaluation results are improved, but also a quantitative basis is provided for the standard establishment, process optimization and market supervision of sea cucumber peptide products, enhancing the applicability and industrial promotion value of the entire system. BRIEF DESCRIPTION OF THE DRAWINGS

[0039] Figure 1 Schematic diagram of the method of the present invention;

[0040] Figure 2 It is a structural schematic diagram of the present invention. DETAILED DESCRIPTION

[0041] The following will clearly and completely describe the technical solutions in the embodiments of the present invention in conjunction with the accompanying drawings. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without making any creative efforts shall fall within the scope of protection of the present invention.

[0042] Example 1

[0043] See also Figure 1 As shown, the method for determining and standardizing the quality of sea cucumber peptides described in the embodiment of the present invention includes the following steps:

[0044] S1: The sample area determined in the spectral absorption diagram and microstructure image of the sea cucumber peptide sample is divided into multiple analysis windows, and the spectral response data of the corresponding position of the analysis window in the spectral absorption diagram is processed by derivative operation to construct a change gradient model.

[0045] The derivative operation is performed in S1 to construct the change gradient model as follows:

[0046] The sample area is divided into N small windows of the same size, numbered as W1, W2, ..., W N , each window W i Has a unique spatial coordinate (x i ,y i );

[0047] For each window W i Extract the corresponding spectral absorption curve, the formula is:

[0048] S i (λ)=[S i (λ1),S i (λ2),...,S i (λ m )]

[0049] Where λ j is the jth wavelength point, m is the total number of wavelength points, S i (λ) is the spectrum curve composed of the absorption values of the i-th window at all wavelengths;

[0050] For each spectral curve S i (λ) is processed by first-order derivative to detect the absorption change trend and improve the sensitivity to small changes. The Savitzky–Golay smoothing derivative algorithm is used, and the formula is:

[0051]

[0052] Where S'k (λ j ) is the wavelength of the i-th window j The derivative value, c k is the Savitzky–Golay smoothing coefficient, n is the half-width of the smoothing window;

[0053] Each window W i The rate of change G i Mapping back to the original image space coordinates (x i ,y i ), construct a two-dimensional changing gradient model.

[0054] It should be noted that the purpose of performing derivative operations and constructing a gradient model is:

[0055] Function 1: Derivative operation highlights subtle changes in the spectral curve, such as peaks and inflection points, suppresses baseline drift and smoothes background noise, helps to discover chemical components or structural features that are not obvious in the original absorption curve, and improves the ability to identify subtle differences in sea cucumber peptides.

[0056] Function 2: Map the rate of change of each analysis window back to the spatial position to form a change gradient model, that is, a two-dimensional graph reflecting the intensity of the change in spectral response. The model can reflect the spatial uniformity and local differences in the spectral response of the entire sample area, providing a spatial distribution basis for subsequent analysis and supporting regional screening and quantitative analysis.

[0057] S2: Identify abnormal areas in the spectral curve based on the variation gradient model, and remove analysis windows corresponding to the abnormal areas in the sample area to obtain candidate analysis areas.

[0058] The spectral curve comes from the collection of the spectral absorption diagram of the sea cucumber peptide sample. The spectral absorption diagram is obtained by measuring the sea cucumber peptide sample using a spectral analysis instrument and a UV-visible spectrophotometer. The spectral absorption diagram records the spectral response of the sea cucumber peptide sample at different wavelengths, that is, the absorbance. The absorbance value at each wavelength point constitutes a data point of the spectral curve.

[0059] The analysis window process corresponding to removing abnormal areas within the sample area in S2 is as follows:

[0060] Based on the constructed variation gradient model, a statistical judgment method is used to identify spectral abnormal areas, and the corresponding analysis windows are eliminated in the sample area to obtain candidate analysis areas. The specific steps include:

[0061] The average change rate and standard deviation are calculated based on the spectral derivative change rate of all analysis windows, and the judgment threshold for spectral abnormality is set. The judgment threshold is the weighted calculation value of the window change rate mean and standard deviation. The formula is:

[0062]

[0063] Where, is the mean of the window change rate in the entire region, σ G is the standard deviation of the rate of change, α is the sensitivity adjustment parameter;

[0064] In some preferred embodiments, the spectral change rate of each analysis window is compared with a threshold value;

[0065] If the spectrum change rate is less than or equal to T, the window is judged to be a normal window;

[0066] If the spectrum change rate is greater than T, the window is determined to be an abnormal window.

[0067] The identified abnormal windows are removed from the sample area, and the remaining windows are retained to form the candidate analysis area. The candidate area is further subjected to spatial connectivity judgment and boundary smoothing processing to remove isolated small blocks or edge outlier windows, thereby improving the spectral stability and representativeness of the candidate area.

[0068] The technical solution of this embodiment is: divide the spectral absorption graph of the sea cucumber peptide sample into windows, extract the spectral response data of each window, calculate the spectral change rate through derivative operation, and construct a change gradient model that represents the intensity of the spectral response; set the change rate abnormality threshold according to statistical characteristics, determine the analysis window with a change rate exceeding the threshold as an abnormal area and eliminate it, and finally retain the remaining windows to form a candidate analysis area with stable spectral response. Through this technical solution, it is possible to effectively exclude spectral abnormal areas caused by impurity interference, abnormal composition or uneven structure in the sea cucumber peptide sample, significantly improve the composition consistency and data stability of subsequent regional analysis, and provide a high-quality and repeatable analysis basis for the construction of the sea cucumber peptide quality grading model, thereby improving the accuracy and standardization of quality evaluation.

[0069] Example 2

[0070] like Figure 1 As shown, a method for determining and standardizing the quality of sea cucumber peptides further includes the following steps:

[0071] S3: Through component deviation analysis, determine whether the chemical composition characteristics of the candidate analysis area are similar to the main component characteristics of the sea cucumber peptide sample; if similar, extract the corresponding microstructure image data in the candidate analysis area and calculate the particle morphology characteristics.

[0072] The process of determining whether the chemical component characteristics of the candidate analysis area are similar to the main component characteristics of the sea cucumber peptide sample in S3 is as follows:

[0073] Based on the spectral response data of sea cucumber peptide samples, principal component analysis was used to extract the principal component eigenvector, which was used as a reference template for chemical composition determination. For multiple standardized sea cucumber peptide samples, full-band spectral absorption data were collected to form the sample spectral data matrix X∈R n*m , where n is the number of samples and m is the number of spectral channels; the spectral data is standardized or normalized to eliminate the measurement error and non-essential differences between samples, and the preprocessed spectral matrix X' is obtained; the matrix X' is subjected to principal component analysis to solve the covariance matrix and extract the first several principal components. Let the principal component loading matrix be P∈R m*k , the principal component score matrix is T∈R n*k , where k is the number of principal components retained; the principal component eigenvector is obtained by multiplying the spectral matrix with the principal component loading matrix. The obtained principal component eigenvector is used as a reference template for subsequent judgment of whether the chemical composition characteristics of the candidate analysis area are similar, and is used for similarity comparison and deviation analysis with the spectral vector of the candidate area.

[0074] Aggregate and extract features from the spectral data of each analysis window in the candidate analysis area to form a feature vector of the candidate area;

[0075] Input the aggregated spectral feature vector of the candidate analysis area and the principal component feature template vector of the sea cucumber peptide sample, and input the two vectors into the similarity analysis model. The indicators include but are not limited to:

[0076] Euclidean distance: represents the absolute distance between two vectors in the feature space. The smaller the distance, the more similar they are.

[0077] Cosine similarity: The value range is [0,1]. The closer the value is to 1, the smaller the angle is and the higher the similarity is.

[0078] Spectral angle mapping: Use angle to represent the similarity of spectral shapes. The smaller the angle, the more similar they are.

[0079] In some preferred embodiments, the spectral feature vector F of the candidate analysis area is obtained. candidate and the principal component template vector F main Then, according to the Euclidean distance D E Compare with the threshold value:

[0080] D E =||F candidate -F main ||2

[0081] If the Euclidean distance is less than the threshold, it is considered similar and the candidate region is retained;

[0082] If the Euclidean distance is greater than or equal to the threshold, it is determined to be dissimilar and the candidate region is eliminated.

[0083] S4: Determine whether the particle morphological characteristics meet the set evaluation requirements. If so, define the candidate analysis area as the effective analysis area. Based on the spectral response data and particle morphological characteristics of the effective analysis area, construct a quality grading model, output the comprehensive quality score, and generate the quality standardization vector of the sea cucumber peptide sample.

[0084] The process of judging whether the particle morphology characteristics meet the set evaluation requirements in S4 is as follows:

[0085] Extract the original image data from the corresponding microstructure image in the candidate analysis area, perform image preprocessing operations such as grayscale conversion, filtering, and binarization to obtain clear particle outlines;

[0086] An image segmentation algorithm is used to extract particle regions and number each particle. The input image is grayscaled and background noise is reduced through Gaussian filtering, median filtering, etc. Histogram equalization is used to enhance particle edge contrast to improve segmentation accuracy. The image is converted into a binary image based on a preset global or local adaptive threshold algorithm to effectively separate the particle region from the background region. The Sobel edge detection operator is used to identify particle boundaries, and the edge results are corrected in combination with morphological operations. Through connected domain analysis, all unconnected particle regions in the image are extracted and pixel boundary information is recorded. A unique number is assigned to each connected region and a label map is generated so that each particle has a clear region index in the image. The uniqueness of the number is verified by traversing and checking whether the maximum value in the label mapping matrix is consistent with the total number of particles. The morphological indicators of all numbered particles are statistically analyzed, including but not limited to particle area, perimeter, aspect ratio, roundness, edge complexity, particle density and distribution uniformity, for subsequent quantitative analysis of quality evaluation.

[0087] The morphological features of multiple particles in the region are aggregated by statistical average, standard deviation, range, etc. to form the morphological feature vector of the candidate analysis area;

[0088] Compare with the preset morphological evaluation requirements to determine whether all morphological indicators meet the set conditions;

[0089] If all morphological indicators meet the set conditions, the candidate analysis area is determined to be a qualified area;

[0090] If any of the morphological indicators does not meet the set conditions, the candidate analysis area is judged as an unqualified area.

[0091] The process of generating the quality standardization vector of the sea cucumber peptide sample in S4 is as follows:

[0092] For the regional data identified as the effective analysis area, the spectral response curve and particle morphological parameters are extracted respectively, and all features are normalized and spliced into a fusion feature vector;

[0093] Based on a large number of standard samples with known quality levels, we extract fusion feature vectors and their corresponding true quality scores, construct a training set, and use a supervised learning algorithm to model the sample data to obtain a scoring model that outputs a comprehensive quality score for unknown samples.

[0094] The fusion feature vector F extracted from the effective analysis area of the sea cucumber peptide sample to be tested test Input model M, output comprehensive quality score Q test =M(F test ), the score is usually normalized to a numerical range of 0 to 100, reflecting the overall quality level of the sample;

[0095] According to the output comprehensive quality score and the corresponding weight contribution of each feature in the model, a quality normalization vector is constructed:

[0096] V std =(w1,f1,w2,f2,...,w n ,f n, Q test )

[0097] Where, f i is the normalized value of the i-th feature in the fusion vector, w i is the importance coefficient of the i-th feature learned in model training, Q test The final calculated comprehensive quality score is used as the final quality representation component in the normalized vector;

[0098] The obtained standardized quality vector is used as a digital expression of the quality of the sea cucumber peptide sample and is used in subsequent processes such as product quality comparison, batch evaluation, production process adjustment or quality traceability analysis.

[0099] The technical solution of this embodiment is as follows: for the candidate analysis area that has passed the composition deviation analysis, the particle area in the extraction is extracted using the image segmentation algorithm, and each particle is numbered and the boundary is identified, and the morphological characteristic parameters of the particles are calculated; the extracted various morphological parameters are compared with the preset evaluation requirements one by one. If all morphological characteristic indicators meet the set qualification standards, the candidate analysis area is confirmed as a valid analysis area; based on the spectral response data of the valid analysis area and the corresponding particle morphological characteristic information, a sea cucumber peptide quality grading model is constructed by fusion, and a comprehensive quality score for reflecting the quality level of the sample is output, and a quality standardization vector with a universal expression is further generated for subsequent product comparison, classification and quality control management. The fusion evaluation of microstructure morphology and spectral response characteristics is effectively realized, so that the quality assessment of sea cucumber peptide samples no longer relies on a single dimensional indicator, thereby improving the scientificity and accuracy of quality grading; at the same time, through the output of the standardized vector, not only the reusability and traceability of the sample quality evaluation results are improved, but also a quantitative basis is provided for the standard establishment, process optimization and market supervision of sea cucumber peptide products, thereby enhancing the applicability and industrial promotion value of the entire system.

[0100] Example 3

[0101] See also Figure 2 As shown, the system for determining and standardizing the quality of sea cucumber peptides according to the embodiment of the present invention includes:

[0102] A system for determining and standardizing the quality of sea cucumber peptides, applied to a method for determining and standardizing the quality of sea cucumber peptides as claimed in any one of claims 1 to 6, characterized in that it comprises:

[0103] Multi-analysis window construction module: multi-window division is performed on the spectral absorption map of the sea cucumber peptide sample and the sample area selected in the microstructure image, and derivative operation is performed on the spectral response data;

[0104] Abnormal area identification module: identifies abnormal areas in the spectral curve based on the change gradient model, and eliminates the corresponding analysis window to obtain the candidate analysis area;

[0105] Morphological feature analysis module: determines whether the chemical composition characteristics of the candidate analysis area are similar to the main component, and extracts the particle morphological characteristics when the conditions are met;

[0106] Quality modeling module: Based on the spectral response data and particle morphological characteristics of the effective analysis area, a quality grading model is constructed, which outputs a comprehensive quality score and generates a quality standardization vector.

[0107] It should be noted that, in this document, relational terms such as first and second, etc., are used only to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply any actual relationship or order between these entities or operations. Moreover, the terms "comprises," "comprising," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus that includes a list of elements includes not only those elements but also other elements not explicitly listed, or elements inherent to such process, method, article, or apparatus.

[0108] While embodiments of the present invention have been shown and described, it will be appreciated by those skilled in the art that various changes, modifications, substitutions, and variations may be made to these embodiments without departing from the principles and spirit of the invention, and that the scope of the invention is defined by the appended claims and their equivalents.

Claims

1. A method for determining and standardizing the quality of sea cucumber peptides, comprising the following steps: The sample area determined in the spectral absorption map and microstructure image of the sea cucumber peptide sample is divided into multiple analysis windows, and the spectral response data of the corresponding position in the spectral absorption map of the analysis window is processed by derivative operation to construct a change gradient model; Identify abnormal areas in the spectral curve based on the variation gradient model, and remove the analysis window corresponding to the abnormal area in the sample area to obtain the candidate analysis area; Through component deviation analysis, it is determined whether the chemical composition characteristics of the candidate analysis area are similar to the main component characteristics of the sea cucumber peptide sample; If similar, the corresponding microstructure image data in the candidate analysis area is extracted and the particle morphology characteristics are calculated; Determine whether the particle morphological characteristics meet the set evaluation requirements. If so, define the candidate analysis area as the effective analysis area. Based on the spectral response data and particle morphological characteristics of the effective analysis area, construct a quality grading model, output the comprehensive quality score, and generate the quality standardization vector of the sea cucumber peptide sample.

2. The method for determining and standardizing the quality of sea cucumber peptides according to claim 1, wherein the derivative operation is performed to construct a gradient model process: The sample area is divided into small windows of uniform size and numbered respectively. Each window has a unique spatial coordinate. The corresponding spectral absorption curve is extracted for each window, and the first-order derivative processing is performed on each spectral curve using the Savitzky–Golay smoothing derivative algorithm; The rate of change of each window is mapped back to the original image space coordinates to construct a two-dimensional change gradient model.

3. A method for determining and standardizing the quality of sea cucumber peptides according to claim 2, characterized in that the analysis window process corresponding to the abnormal area is eliminated in the sample area: The spectral curve comes from the collection of the spectral absorption diagram of the sea cucumber peptide sample, and the spectral absorption diagram is obtained by measuring the sea cucumber peptide sample; Based on the constructed variation gradient model, a statistical judgment method is used to identify spectral abnormal areas and eliminate the corresponding analysis windows in the sample area; The average change rate and standard deviation are calculated based on the spectral derivative change rate of all analysis windows, and the judgment threshold for spectral change anomaly is set. The judgment threshold is the weighted calculation value of the window change rate mean and standard deviation.

4. A method for determining and standardizing the quality of sea cucumber peptides according to claim 3, characterized in that the process of determining whether the chemical composition characteristics of the candidate analysis area are similar to the main component characteristics of the sea cucumber peptide sample is: Based on the spectral response data of sea cucumber peptide samples, principal component analysis was used to extract the principal component eigenvectors, which were used as reference templates for chemical composition determination. Aggregate and extract features from the spectral data of each analysis window in the candidate analysis area to form a feature vector of the candidate area; The aggregated spectrum feature vector of the candidate analysis area and the principal component feature template vector of the sea cucumber peptide sample are input, and the two vectors are input into the similarity analysis model.

5. A method for determining and standardizing the quality of sea cucumber peptides according to claim 4, characterized in that the process of determining whether the particle morphology meets the set evaluation requirements is: Extracting raw image data from the corresponding microstructure image in the candidate analysis area, performing preprocessing operations, and obtaining particle contours; Use image segmentation algorithm to extract particle area and number each particle; Aggregate the morphological features of multiple particles in the region to form a morphological feature vector of the candidate analysis region; Compare with the preset morphological evaluation requirements to determine whether all morphological indicators meet the set conditions.

6. The method for determining and standardizing the quality of sea cucumber peptides according to claim 5, wherein the process of generating the quality standardization vector of the sea cucumber peptide sample is: For the regional data identified as the effective analysis area, the spectral response curve and particle morphological parameters are extracted respectively, and all features are normalized and spliced into a fusion feature vector; Extract the fusion feature vector and the corresponding true quality score, build a training set, and use a supervised learning algorithm to model the sample data to obtain a scoring model; The fusion feature vector extracted from the effective analysis area of the sea cucumber peptide sample to be tested is input into the model, and the comprehensive quality score is output; A quality normalization vector is constructed based on the output comprehensive quality score and the corresponding weight contribution of each feature in the model.

7. A system for determining and standardizing the quality of sea cucumber peptides, applied to a method for determining and standardizing the quality of sea cucumber peptides as claimed in any one of claims 1 to 6, characterized in that it comprises: Multi-analysis window construction module: multi-window division is performed on the spectral absorption map of the sea cucumber peptide sample and the sample area selected in the microstructure image, and derivative operation is performed on the spectral response data; Abnormal area identification module: identifies abnormal areas in the spectral curve based on the change gradient model, and eliminates the corresponding analysis window to obtain the candidate analysis area; Morphological feature analysis module: determines whether the chemical composition characteristics of the candidate analysis area are similar to the main component, and extracts the particle morphological characteristics when the conditions are met; Quality modeling module: Based on the spectral response data and particle morphological characteristics of the effective analysis area, a quality grading model is constructed, which outputs a comprehensive quality score and generates a quality standardization vector.

Images